The objective of this research is to provide a quantitative framework for researchers and policy makers developing prevention and screening strategies to control prostate cancer. Research into prostate cancer prevention is expanding and maturing. The Prostate Cancer Prevention Trial (PCPT), a phase III trial of finasteride as a possible preventive agent, is scheduled for completion in 2004. A second Phase III prostate cancer prevention trial is currently under way to test the efficacy of selenium and vitamin E in preventing the disease. At the same time, Prostate-Specific Antigen (PSA) screening has become common practice in the US. Data on PSA utilization and prostate cancer incidence over the past decade are consistent with rapid dissemination of the test for early detection purposes. How should limited health care dollars be allocated when both screening and prevention are available for control of prostate cancer? The central theme of this proposal is that computer modeling provides an appropriate and powerful means to answer this important and timely question. Our primary Specific Aim is to develop a comprehensive computer model of prostate cancer prevention and screening at the level of the individual. The model will add a primary prevention component and an economics front end to an established model of prostate cancer natural history and PSA screening developed by the study investigators. For any strategy combining a specific screening schedule with a primary preventive agent, the model will project the years of life saved, the costs of the strategy less any savings in treatment costs, and the incremental cost per year of life saved. Our second Specific Aim is to use the model together with data from the PCPT to project the incremental costs and benefits of different strategies combining finasteride with PSA screening. In addition to illustrating the utility of our model, this aim will also allow us to estimate the cost-effectiveness of finasteride as a preventive agent for prostate cancer. As part of this aim, we will use the PCPT data to estimate the effects of finasteride on disease natural history, including disease onset, tumor metastasis, and PSA growth rates. The modeling infrastructure developed through the proposed work will be applicable to any potential preventive agent. The final model will be made available to investigators through a dedicated website which will allow users to run the model with their own settings for key cost and efficacy parameters. [unreadable] [unreadable]